To explain the difficulty that managers face in identifying explorative opportunities, extant literature highlights a top-down attentional process in which cognitive representations have a filtering effect on searching and interpreting data. However, the inherent power of data to update cognitive representations and enable the identification of explorative opportunities has received less attention. This study develops a new data-centered logic, complementary to predominant cognitive representation-centered explanations, by investigating how data-driven strategic decision-making affects exploration and exploitation. Based on the attention-based view, we propose that data-driven strategic decision-making will induce three interaction processes—self-justification, adaptive reflection, and revolutionary reflection. We employ these processes to reveal the U-shaped effect of data-driven strategic decision-making on exploration and the positive effect on exploitation. Besides, we also propose that paradoxical cognition flattens the U-shaped effect, whereas it strengthens the positive effect. We employed survey data on 334 manufacturing firms to examine hypotheses, and the robustness of findings is also supported by supplementary interviews with 8 firms. Our findings indicate that cognitive representations do not necessarily lead to significant filtering effects if managers employ high levels of data-driven strategic decision-making. Such decision-making can leverage the inherent power of data to change cognitive representations, break the filtering effects, and then promote exploration for new business opportunities.
Technological innovation in the semiconductor industry now unfolds under extreme process–design coupling, capital intensity, and compressed product cycles, making traditional interface-based collaboration insufficient. This study addresses three questions: (RQ1) under what conditions does deeply integrated co-creation emerge? (RQ2) through what mechanisms does deep integration accelerate innovation? and (RQ3) what governance and resilience trade-offs does it generate? Using a longitudinal, inductive, theory-building case study of NVIDIA and TSMC, we triangulate secondary evidence from financial indicators, technology milestones, and industry reports. We develop a conceptual model of “deeply integrated co-creation”—defined as rapid, non-separable, process-level co optimization across firm boundaries supported by governance, IP safeguards, and relational routines—and introduce “strategic symbiosis” as its relational structure. Rather than proposing a universally superior model, this study specifies when and why deeply integrated co-creation becomes necessary under extreme technological uncertainty, asymmetric capital intensity, and non separable design–process coupling. The analysis yields three theoretically grounded insights: (1) asymmetric CAPEX–R&D specialization may function as dynamic complementarity accelerating time-to-market; (2) deep integration emerges when non modularity makes specification-only coordination infeasible; and (3) resilience trade-offs increase with concentration, requiring managerial and policy safeguards. The resulting framework is intended as an analytically generalizable theory-building model whose broader applicability remains subject to future comparative validation.
Industry 5.0 (I5.0) advances the shift from Industry 4.0's (I4.0) technology-driven industrial models to systems that prioritize sustainability, human-centricity, and resilience. Its implementation remains limited, however, due largely to insufficient empirical evidence on the organizational barriers that impede adoption, particularly in environments shaped by continuous improvement practices. This study identifies and prioritizes the key barriers to I5.0 implementation and examines how organizations with differing levels of Lean Six Sigma (LSS) maturity perceive their readiness to engage with this emerging paradigm. Using mixed-methods design that integrates three methods (a two-round Delphi study with 19 experts, a systematic literature review, and comparative case analyses of two companies), the study identifies 15 critical barriers. The most significant relate to strategic misalignment, cultural resistance to change, limited leadership engagement, and lack of structured pathways for socio-technical integration. The results indicate that the LSS-mature firm demonstrates stronger readiness across resilience, strategic alignment, and process integration, whereas the non-LSS firm appears to show lower perceived preparedness across all I5.0 pillars. To support implementation, the study maps specific LSS tools to each barrier, providing an actionable framework to diagnose obstacles, guide transformation initiatives, and integrate operational excellence with I5.0 objectives. Rigor was strengthened through methodological sequencing, expert-informed validation and prioritization across Delphi rounds, and cross-case comparison. By linking theoretical insights to practical guidance, the study strengthens the foundation for I5.0 as a pragmatic, implementable model for sustainable, human-centered, resilient industrial development.
As consumer data becomes an increasingly strategic asset in digital commerce, competing online sellers face critical trade-offs in business mode choices. This study develops a game theoretic model to investigate whether competing online sellers should adopt the commission mode or the data monetization mode. Our results show that large sellers with advanced data analytics capability should purchase more data under both symmetric and asymmetric channel structures, while small sellers should purchase less due to constrained data analytics capability. Interestingly, the channel structure significantly influences outcomes; in asymmetric channels, smaller sellers can sometimes achieve higher demand and profits despite their disadvantages. Our analysis further reveals a misalignment between sellers' and the platform's preferences; sellers prefer the commission mode but hope their competitors choose the data monetization mode. In contrast, the platform consistently favors the commission mode for large sellers, and its preference for small sellers depends on misfit cost. As robustness checks, we extend the baseline model by considering further scenarios including partial data sharing and endogenous data pricing. Based on these extensions, deeper insights into effective data management strategies for online platform operations are offered.
Semiconductor supply chains (SCs) depend on geographically concentrated rare earth element inputs and face systemic risks from geopolitical disruption, policy intervention, and capacity constraints across multiple decision-making layers. Existing stress testing approaches often quantify fulfilment loss without attributing it to strategic capacity shortages, tactical sourcing-access constraints, or operational congestion, even though each requires a different intervention. This study develops a hybrid simulation modelling framework integrating System Dynamics (SD), Agent-Based Modelling (ABM), and Discrete Event Simulation (DES) as necessary analytical approaches per decision-making layer: SD captures long-term capacity stocks and policy feedback (macro-level); ABM captures heterogeneous monthly sourcing and inventory decisions (meso-level); and DES captures stochastic weekly fabrication and shipment execution (micro-level). Applied to India's cerium-centric semiconductor SC through 247 experiments, 12,350 simulation runs and an additional 1,400 runs of sensitivity analysis, the results show that upstream cerium capacity disruptions are largely buffered, whereas loss of qualified fabrication capacity causes severe fulfilment collapse. Stable DES cycle times further confirm that this failure arises from sourcing access rather than operational congestion. This cross-layer diagnostic signal is only detectable when all three modelling layers are present. Managerially, the hybrid simulation modelling framework reframes stress testing as a multi-level tool that identifies not only whether an SC fails, but also where failure emerges and which interventions are appropriate.
